Hugging Face's logo

Spaces:
maladroitcat
/
semantic_search
Sleeping

App Files Community
semantic_search / README.md
Abby Reynolds
Readme.md formatted for hf spaces.
41ab2f9
preview code
|
raw
history blame contribute delete
977 Bytes

A newer version of the Gradio SDK is available: 6.1.0

Upgrade
metadata
title: semantic_search
emoji: 🔍
colorFrom: blue
colorTo: pink
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false

Semantic Search (CNN/DailyMail demo)

Gradio app that builds a Chroma vector store from a small slice of the CNN/DailyMail dataset, then serves semantic search using a SentenceTransformer encoder.

Run locally

  1. python -m venv .venv && source .venv/bin/activate
  2. pip install -r requirements.txt
  3. python app.py
    • First run downloads a 100-row slice of abisee/cnn_dailymail, chunks articles (500 words, 50 overlap), embeds with all-MiniLM-L6-v2, and persists to chromadb/.
    • Visit the Gradio UI at http://localhost:7860.

Deploy on Hugging Face Spaces

  • Space type: Gradio (Python).
  • Push app.py and requirements.txt (the chromadb/ folder is created at runtime).
  • The app binds to PORT if provided by the Space; defaults to 7860.

DISCLAIMER: This README.md was formatted with help from ChatGPT